Active disturbance rejection control of a parallel manipulator with self learning algorithm for a pulsating trajectory tracking task

A novel and robust intelligent scheme is proposed to control a highly non-linear 3-RRR (revolute-revolute-revolute) planar parallel robotic manipulator, via an Active Force Control (AFC) strategy that is embedded into the classic Proportional-Integral-Derivative (PID) control loop. A PID-type Iterative Learning (IL) algorithm, with randomized initial conditions, is incorporated into the AFC loop to approximate the estimated inertia matrices of the manipulator adaptively while the manipulator is tracking a prescribed pulsating trajectory in the presence of harmonic disturbances. The IL algorithm employs a stopping criterion, which is based on tracking error, to stop the learning process when the desired error goal of the system is reached, to signify a favorable controlled condition. A numerical simulation study was performed to verify the robustness of the proposed methodology in rejecting disturbances, based on given loading and operating environments. The results of the study reveal the superiority of the proposed system, in terms of its excellent tracking performance compared to the AFC, with crude approximation techniques, and Proportional-Integral-Derivative (PID) counterparts.